How to Report Chi-Square Results in APA Format

APA Reporting Template

Use this template to report your chi-square results. Replace the bracketed placeholders with your values.

Chi-Square Test of Independence

A chi-square test of independence was performed to examine the relationship between [variable 1] and [variable 2]. The relation between these variables was significant, $\chi^2$([df], N = [total N]) = [chi-square value], p = [p-value], [Cramer's V / $\phi$] = [value].

Chi-Square Goodness-of-Fit Test

A chi-square goodness-of-fit test indicated that the distribution of [variable] was significantly different from [expected distribution], $\chi^2$([df], N = [total N]) = [chi-square value], p = [p-value].

Worked Example

Scenario: A researcher examined whether treatment completion (completed vs. dropped out) was related to therapy type (CBT, psychodynamic, or medication only) among 180 patients.

Results:

• $\chi^2(2, N = 180) = 9.87, p = .007$
• Cramer's $V$ = .23

APA Write-Up:

A chi-square test of independence was performed to examine the relationship between therapy type and treatment completion. The relation between these variables was significant, $\chi^2$(2, N = 180) = 9.87, p = .007, Cramer's V = .23, indicating a small-to-medium association. Patients in the CBT condition were most likely to complete treatment (83%), followed by the medication-only condition (72%), and the psychodynamic condition (62%).

Reporting Checklist

• [ ] Named the type of chi-square test (independence or goodness-of-fit)
• [ ] Stated both categorical variables
• [ ] Reported $\chi^2$ with degrees of freedom and sample size: $\chi^2$(df, N = [value])
• [ ] Reported the exact p-value (or p < .001)
• [ ] Included an effect size measure (Cramer's V for tables larger than 2 x 2, $\phi$ for 2 x 2 tables)
• [ ] Reported observed frequencies or percentages for each cell
• [ ] Noted whether any expected cell frequencies were below 5
• [ ] Used italics for statistical symbols (N, p, V)

Common Mistakes

1. Omitting sample size — Always include N inside the parentheses: $\chi^2$(2, N = 180), not just $\chi^2$(2).
2. Using the wrong effect size — Use $\phi$ (phi) for 2 x 2 tables and Cramer's V for larger tables. They are identical for 2 x 2 tables, but Cramer's V is the appropriate generalization for larger tables.
3. Not reporting frequencies — The chi-square statistic alone does not tell the reader what the pattern looks like. Always include observed counts or percentages.
4. Ignoring the expected frequency assumption — If more than 20% of expected cell frequencies are below 5, report this and consider using Fisher's exact test instead.
5. Treating chi-square as a measure of strength — A large $\chi^2$ does not necessarily mean a strong association; it can be inflated by large sample sizes. Always report Cramer's V or $\phi$ to convey effect size.
6. Writing "p = .000" — Write p < .001 instead.

Non-Significant Results

If your chi-square test is not significant, still report all the same information:

A chi-square test of independence was performed to examine the relationship between therapy type and treatment completion. The relation between these variables was not significant, $\chi^2$(2, N = 180) = 2.14, p = .343, Cramer's V = .11. Treatment completion rates were similar across CBT (75%), psychodynamic (70%), and medication-only (68%) conditions.

Results Table Format

Report a contingency table with observed frequencies and percentages:

For tables with many variables, also consider reporting standardized residuals to show which cells deviate most from expected values: